Skip to content

Siphon - Effortlessly extract, compress, and cache Git repository contexts for integration with LLMs.

License

Notifications You must be signed in to change notification settings

atxtechbro/siphon

Repository files navigation

Siphon

Efficiently extract, compress, and cache Git repository contexts for seamless integration with Large Language Models (LLMs).

License Python Versions PyPI Build Status Coverage Downloads Issues

Funded by YC (Your Cents) 😉


Table of Contents


Features

  • Efficient Extraction: Extracts and compresses repository contents while respecting .gitignore rules.
  • Customizable Filtering: Include or exclude files and directories with ease.
  • Multiple Output Formats: Supports text, tarball, and markdown formats optimized for LLM contexts.
  • Caching and Chunking: Pre-cache large repositories for faster querying.
  • Token Count Estimations: Get token counts for specific LLMs like GPT-3 and Claude.
  • Clipboard and Stdout Support: Streamline workflows with seamless copying options.
  • Modularity: Extend functionality with community-driven extensions.
  • Interactive Mode: Granular file selection through an interactive interface.

Installation

Install Siphon using pip:

pip install siphon-cli

Usage

Navigate to your Git repository and run:

si -o context.txt

This command extracts the repository content into context.txt.


Examples

  • Include Specific File Types:

    si -i "*.py" -o python_files.txt
  • Exclude Directories:

    si -e "tests/*" -o code_without_tests.txt
  • Interactive Mode:

    si --interactive -o selected_files.txt
  • Copy Output to Clipboard:

    si --clipboard

Arguments

  • path: Path to the Git repository (default: current directory).
  • -i, --include: Include file patterns (e.g., .py, src/).
  • -e, --exclude: Exclude file patterns (e.g., tests/, *.md).
  • -o, --output: Output file name (default: output.txt).
  • -f, --format: Output format (text, tar, markdown).
  • -c, --cache: Enable caching (future feature placeholder).
  • --tokenizer: Tokenizer for token count estimation (gpt3, claude).
  • --interactive: Interactive mode for file selection.
  • --clipboard: Copy output to clipboard.
  • --stdout: Print output to stdout.

Contributing

We welcome contributions from the community! To contribute:

  1. Fork the repository.

  2. Create a new branch:

    git checkout -b feature/your-feature-name
  3. Commit your changes:

    git commit -am 'Add a new feature'
  4. Push to the branch:

    git push origin feature/your-feature-name
  5. Open a Pull Request.

Please read our Contributing Guidelines for more details.


License

This project is licensed under the MIT License.


Contact